News for hackers. Stuff that matters.

vanagandr

How do you stay out of it all, if at all?
Re:So you want to use plants to reduce CO₂ on Friday July 31, 2026 @12:52AM (#49119144)
I can’t find a single credential or scientific qualification of merit of the author and yet their commentary spans many topics, including cancer.

I’ll pass. The About page is not very helpful either. It says very little about the author except this:

> Like most people, I use AI “for research”, but I impose fairly strict limits to make sure I don’t use AI “for writing”.

Re:The Cold Email on Wednesday July 29, 2026 @10:52PM (#49105623)
Nice to see Zach still posting.

I have had reasonable success with cold emails. My favorite genre is every so often I email a musician or author whose works I tremendously enjoyed asking for a signed copy, an opinion, or simply to say thank you for having a profound influence on me. Some are very well known. But almost all reply with enthusiasm.

Re:Mitchellh starts a new company: Superlogical on Wednesday July 29, 2026 @12:44PM (#49099840)
If its got Mitchell, I’m in.

To make it more relevant: right now I mostly rely on tmux. Before that I’ve been a long time user of screen (irssi over ssh people know this play). Looking forward to this.

Re:AI in Linux on Wednesday July 29, 2026 @12:37PM (#49099727)
This is a “everything is political” rant but now with 50% more AI.
Re:Windows 11 is quietly installing OneDrive Photos on Wednesday July 29, 2026 @11:21AM (#49098707)
I’ve switched over to Windows 2021 Enterprise IoT LTSC before I jump to Linux. Just letting it slow burn now.
Re:I'm 38 and I Can't Support Myself Anymore on Wednesday July 29, 2026 @01:36AM (#49093778)
What kind of work do you do right now?
Re:A Deep Logo Study on Tuesday July 28, 2026 @01:23PM (#49087108)
One thing AI will reveal is a lack of prose in people’s writing. The way most of this post is written is, well, kind of stilted. I don’t have a better word for it. Some people read it and equate to an LLM output.

The way it ends is all over the place:

“I have a name in my head for the repair side, and two cameras on the desk, and no idea yet whether any of it becomes a real thing. A mute mark does not force me to decide. It works as the logo of a freelance developer today, it would work on a box tomorrow, and it does not look silly either way.

Which is roughly what the component does.

It holds because it is not closed.”

Re:Our position on open-weights models on Monday July 27, 2026 @11:49PM (#49079101)
> The reality is much less confusing

The reality is even less confusing than that: China is amused by the kvetching tactics. They know who their opponents are but are cunning enough to not reveal their cards.